Abstract
Purpose :
Retinal projections to the dLGN form two distinct types of retinogeniculate synapses (termed "simple" and "complex"), both of which are structurally and functionally unique compared with other retino-recipient nuclei. We identified Neuritin 1 (NRN1) and Leucine Rich Repeat Transmembrane Neuronal 1 (LRRTM1), as two cues that may drive differentiation of retinal axons into "complex" retinogeniculate synapses. Analysis in mice lacking these factors revealed no morphological difference in the absence of NRN1, and a loss of "complex" retinogeniculate synapses in the absence of LRRTM1. Here, we took advantage of the loss of "complex" retinogeniculate synapses in these mutants to assess the functional significance of retinal convergence in dLGN on visual behavior.
Methods :
We evaluated visual behavior of adult nrn1-/- and lrrtm1-/- mice in a Y Water Maze test. In this test, mice are trained to associate a visual cue, a black and white grating (S+), with a hidden platform that allows them to escape the water. Simple aspects of visual behavior were tested by varying the spatial frequency, contrast, or orientation of the grating. Complex visual behavior was tested by simultaneously varying multiple visual features of the grating. Accuracy of lrrtm1-/-, nrn1-/-, and wildtype (WT) control mice in locating the hidden platform were compared using two-way ANOVA in Prism (n=12-20).
Results :
The loss of either LRRTM1 or NRN1 did not result in deficits in performance of simple behavioral tasks. However, ANOVA analysis demonstrated a significant difference among lrrtm1-/-, nrn1-/-, and WT mice in performance of complex behavioral tasks (p<.0001). While our analysis revealed no difference in the ability of control and nrn1-/- mice to discriminate vertical and horizontal bars at a higher spatial frequency of .32 cycles per degree (p>.9), lrrtm1-/- mice accuracy dropped significantly during these more complex tasks in comparison to controls (p<.0001). Similarly, lrrtm1-/- mice lacked the sensitivity to discriminate the two visual patterns at lower contrast (25% p<.0006, and 10% p<.03).
Conclusions :
These results indicate that mice lacking LRRTM1 and "complex" retinogeniculate synapses exhibit abnormalities in more complex visual tasks. These findings provide the first insight into the functional significance of retinal convergence in mouse visual thalamus in the processing and relaying of visual information.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.